3. Regional council methods for monitoring freshwater quality

3.1: Environment Southland

3.1.1: Rivers physical-chemical and microbiological

Baseline river water quality monitoring commenced in Southland in 1989 as part of the National River Water Quality Monitoring Network run by NIWA. There are five sites in the NRWQN within the region, located on four main river systems (Waiau, Aparima, Oreti and Mataura). To supplement this network, Environment Southland established a comprehensive monitoring programme in 1994 and 1995. Regular monthly monitoring of faecal indicator bacteria commenced in July 1994 at 15 sites. In July 1995, a network of 26 sites (not including, the NIWA sites) was established to monitor physico-chemical variables. This network was modified between July 1999 and July 2000 to incorporate tests for faecal indicator bacteria at all water quality sites (Meijer 2010).

Currently, Environment Southland conducts monthly monitoring of 70 sites on 43 rivers (Table 4). The majority have records of 10-15 years of data, providing a good basis for trend analysis. There are 17 geographical zones, but some do not have surface water monitoring sites (viz. Fiordland, Coastal Longwoods, Stewart Island may be included in future following a review by Aquanet), with the reason being that the sites are very difficult to access, or have few pressures on the resources that warrant inclusion in the SoE monitoring programme.

Table 4:

Total number of current river SoE monitoring sites in each region by variables analysed. The number of sites for which the ten water quality variables are analysed for each region is provided in Section 4 for Southland (Table 9), Taranaki (Table 15), Horizons (Table 21) and Waikato (Table 25).

There were 56 SoE water quality monitoring sites in the Southland region that met our criteria for trend analysis. The majority of these sites were located in Low-elevation Topography and Pastoral Land-cover categories. These categories are also the most commonly occurring REC category by river length (Table 5). The Hill Topography category was well represented by 16 sites that were distributed over Exotic and Indigenous Forest, Tussock and Pasture Land-cover categories. The network included only one site in the Mountain Topography category and 4 in the Lake Topography category. Thus, the monitoring network gave good representation of the environmental variability of the region's rivers, but, as expected under-represented unimpacted regions – because of difficulty of access and lack of obvious pressures. Environment Southland staff have recently reviewed their SoE monitoring programme and recommended an increase to 94 water quality monitoring sites. This provides the opportunity to refine the representativeness of the network by including more unimpacted (reference) and urban sites, and modifying the spatial coverage of water quality sites to improve representation of management zones defined by the Regional Water Plan.

Table 5:

Numbers of river SoE water quality monitoring sites in the Southland region that met our criteria for trend analysis and (in brackets) the number of apparent SoE sites in 2009 (i.e. having at least one sample in each season during 2009). The sites have been classified by REC Topography and Land-cover categories (see Table 3) and the colour scale indicates the total proportion of river length in the Region that is classified by each of the categories.

Flow data are available for most occasions when water quality data is obtained; either from permanent flow monitoring sites, actual gaugings, or estimates based on nearby flow monitoring sites. Thirteen sites are at Environment Southland flow gauging stations site; six sites have NIWA gauging stations; two sites have rated Environment Southland gauging stations and 15 have a gauging station within 15 km (total of 36). The remaining 36 sites have flow measurements estimated from known relationships to nearby flow sites. While this means that sites are generally associated with flow data for each monitoring occasion more detailed analysis would be needed to assess whether the estimated flows are adequate or should be improved.

Samples were routinely analysed for 14 water quality variables: black disc water clarity, conductivity, dissolved oxygen, dissolved reactive P, total P, E. coli, faecal coliform, ammonium N, nitrate plus nitrite N, total N, pH, total suspended solids, temperature and turbidity. Samples from seven sites were also monitored for BOD. The range of variables is very similar to that of the NRWQN except that ES do not measure CDOM, but do include SS and FC (as well as E. coli).

Statistically significant differences in water quality state for sites grouped by REC Topography and Land-cover categories were observed for most variables (see Section 4.1.1). In addition, overall trends (in specific REC Topography and Land-cover categories) were observed for several variables (see Section 4.1.2). These analyses indicate that the river water physical-chemical and microbiological quality monitoring program is adequate, at least from the perspective that it has sufficient statistical power (numbers of sites relative to their variability) to detect large scale patterns (as defined by REC categories) in water quality state and trends in the Southland region.

3.1.2: Rivers biological

Invertebrate and periphyton surveys have been conducted at approximately 70 sites per year since 2007. This monitoring was started in 1996, with more sites added over time such that about 70 sites (on average) have been surveyed annually since 2007. Invertebrate data are expressed as MCI and SQMCI scores to detect changes over time for river ecosystem health. Periphyton samples (collected following Stark 2010) are analysed for chlorophyll a and Ash Free Dry Weight (AFDW) and assessed according to standard procedures. Periphyton are sampled once annually during summer, when pressures of temperature and algae growth and are likely to be highest, and river flows are low and stable. Annual biological sampling during summer is a useful way of capturing habitat stress at low flows and high temperatures, but may miss seasonal land use impacts such as the peak production from dairy farming that occurs during spring and is subject to bias from atypical conditions (e.g. higher than usual rainfall and river flow). However, annual sampling is generally considered to be suitable (and is the protocol of the NRWQN, for example) (Stark and Phillips 2009).

3.1.3: Contact recreation and drinking water

Water samples are analysed for E. coli (a faecal indicator organism) from 11 popular freshwater bathing sites over summer. Seven sites are tested weekly between December and March each summer, and four sites are tested monthly over the whole year with the aim being to warn the public of potential health risk. In addition, faecal coliforms have been monitored at monthly intervals at 71 representative river sites throughout the region, since 1994. This second data set provides a picture of trends in faecal contamination within the region; especially relevant given the large number of dairy conversions in Southland. These programmes are augmented by case-study investigations such as a study of faecal pollution in Waituna Lagoon. Bacterial concentrations are evaluated according to the National bathing guideline thresholds (MfE and MoH, 2003).

Drinking water monitoring is predominantly of groundwaters and is evaluated according the the Drinking Water Standard for New Zealand (Ministry of Health). The Standard specifies maximum acceptable values (MAV) for E. coli, nitrate-N, manganese, pH, chloride, total hardness, sulphate, sodium and iron.

3.1.4: Lakes

Environment Southland began regular lake water quality monitoring in July 2000 on Lake Te Anau, and on Lake Manapouri in July 2002. Three small lakes on Southland's southern coast (Lake George, Lake Vincent and The Reservoir) were sampled in early 2002. Prior to this, data on the water quality of Southland's lakes was limited to one-off investigations and spot samples. Regular water quality monitoring on the Waituna Lagoon commenced in October 2001 (Meijer 2010). These lakes are either potentially threatened by runoff from land use intensification, or are actually undergoing change. There are several lakes in the region that are not presently monitored, including Lake Monowai, Lake Hauroko, Lake Poteriteri and Lake Hakapoua – which are all difficult to access but likely to be in pristine condition. However, the inclusion of these large, nationally valuable, lakes in the SoE programme may be worth consideration in future. Environment Southland staff have recently reviewed their SoE monitoring programme and recommended water quality monitoring of more coastal lakes.

Lake waters are sampled at monthly intervals at two sites on Lake Te Anau and three sites on Lake Manapouri, at two depths on occasions when the lake is isothermal (i.e. top and bottom waters are within 3°C), and from the epilimnion and hypolimnion when the lake is stratified. Waituna lagoon is sampled monthly at four sites. Regular monitoring commenced at one site (East) in October 2001 and three additional sites were included (West, Centre, South) in August 2003. Sites are located in the deepest areas of the lagoon near freshwater and saline inputs. Surface water samples (only) are collected and analysed for the same suite as for the lakes, but with the addition of E. coli (a faecal indicator organism).

Lake waters are analysed in the field for: Secchi depth, dissolved oxygen, temperature and conductivity. Samples are analysed in the laboratory for pH, nitrate+nitrite N, ammoniacal N, total N, dissolved reactive P, total P, chlorophyll a, turbidity, total and volatile suspended solids, by RJ Hill Laboratories Ltd using standard methods and procedures for sampling and sample stabilisation. Environment Southland should give consideration to lower detection levels for some variables, particularly when looking at pristine lake water chemistry. The absence of E. coli in the suite of lake water variables is notable but may be included elsewhere in bathing water programmes. Given the changes in land use within Southland it may be appropriate to include E. coli as a water quality variable in the lakes monitoring programme.

3.1.5: Groundwater

Groundwaters are classified into 29 zones that cover the major aquifers of the region and fall into four broad categories: riparian aquifers, terrace aquifers, lowland aquifers and fractured rock aquifers. Groundwater is extensively used for domestic, stock and municipal water supplies throughout the Southland Region. Groundwater is also extensively used for industrial and farm (particularly dairy shed) supply which also require water of potable quality.

The main monitoring network comprises approximately 45 sites distributed across the major aquifer systems that are sampled quarterly for a wide range of variables: conductivity, chloride, sodium, bicarbonate, calcium, magnesium, total hardness, manganese, iron, nitrate, dissolved reactive phosphorus and E. coli (in 200 wells within the main network). In addition, about 250 sites are monitored for resource consent compliance and there are several other ad hoc investigations of groundwater quantity and quality (e.g. nitrate "hotspots", pesticides). Water quality is assessed principally against the Drinking Water Standards for New Zealand (2008) for potable supply, or the ANZECC (2000) guideline for stock water supply or other uses.

3.1.6: QA/QC and data storage

Water chemistry samples are now analysed by RJ Hill Laboratories, Christchurch (IANZ accredited laboratory) with appropriate methods for natural waters (Meijer 2010). It is worth noting that detection levels (DLs) for some variables reported by Hill Laboratories are higher than for other environmental "water" laboratories. For example, DLs cited by Hills for total N and ammoniacal N are 0.1 and 0.01 g m^-3, respectively, compared with 0.01 and 0.001 g m^-3 for the NIWA water laboratory in Hamilton. This will reduce the future ability of the program to detect trends in water bodies that are currently in good condition. Some 'split' samples are sent to Hill Laboratories and to two other laboratories (Environment Canterbury and MLS Envirolab, Invercargill) for interlaboratory comparison. Results showed some variation between laboratories but were generally satisfactory.

Invertebrate samples are analysed by Ryder Consulting, Dunedin with QA/QC check analyses provided by Stephen Moore, Landcare Research, Auckland. The check analyses indicate that most invertebrates were correctly identified by Ryder Consulting and that the numbers of individuals missed by Ryder (but found by Landcare) in the sorted residue were generally low. Environment Southland have recently reviewed their SoE monitoring programme and recommended an increase in the number of annual river biomonitoring sites.

Environment Southland supplied us with multiple spreadsheets and do not appear to have a database for their water quality data. This makes it difficult to restructure data or to efficiently interrogate the data sets.

3.2: Taranaki Regional Council

3.2.1: Rivers physical-chemical and microbiological

Samples from TRC's river SoE monitoring sites are routinely analysed for 19 water quality variables Temp, flow, DO, BOD5, pH, conductivity, black disc visibility, turbidity, absorbance (3 wavelengths), ammonia-N, nitrate-N, total-N, DRP, total P, alkalinity, SS, and faecal coliforms E. coli, and enterococci bacteria. This suite of microbial and physico-chemical variables provides a good description of the land use impacts and the suitability of the monitored water resources for protection of ecosystem health, human contact and recreation and stock health, and provides a sound data base for evaluating river water quality status, trends and loads.

Monitoring sites are distributed across the region and distinguish differences between predominantly forest catchments (near the National Park boundary) and predominantly pasture catchments (often near the coast). However, there are no sites that represent unimpacted (reference) conditions. This seems to be a limitation caused by the inaccessibility of upland reference sites and the extensive nature of intensive agriculture in the region.

There were 12 SoE water quality monitoring sites in the Taranaki region that met our criteria for trend analysis. The majority of monitoring sites were predominantly located in Low-elevation Topography and Pastoral Land-cover categories, which was also the most commonly occurring REC category by river length (Table 6). The indigenous Forest (IF) category of land-cover, particularly in lowlands, appears under-represented. However, the representativeness is a little difficult to assess with only 12 SoE sites.

Table 6:

Numbers of river SoE water quality monitoring sites in the Taranaki region that met our criteria for trend analysis and (in brackets) the number of apparent SoE sites in 2009 (i.e. having at least one sample in each season during 2009). The sites have been classified by REC Topography and Land-cover categories (see Table 3) and the colour scale indicates the total proportion of river length in the Region that is classified by each of the categories.

River flows are estimated in all TRC river water quality monitoring. Flow gaugings are carried out each month at four sites for ratings purposes. Flows are recorded from all other sites but only added to the data base after the site has been validated by TRC or (depending on the site) NIWA hydrological staff and ratings have been verified for the sampling periods. The combination of instantaneous flow-concentration data pairs enables flow-adjustment for trend analysis and for loads to be calculated and compared temporally and spatially.

Despite the small number of SoE sites in the Taranaki region compared to the other regions, statistically significant differences in water quality state for sites grouped by REC Topography and Land-cover categories were observed for some variables (see Section 4.2.1). In addition, a small number of overall trends (in specific REC Topography and Land-cover categories) were observed for several variables (see Section 4.2.2). These analyses indicate that the river water physical-chemical and microbiological quality monitoring program is adequate, at least from the perspective that it has sufficient statistical power (numbers of sites relative to their variability) to detect large scale patterns (as defined by REC categories) in water quality state and trends in the Taranaki region.

3.2.2: Rivers biological

TRC conduct freshwater macroinvertebrate and nuisance periphyton monitoring. Invertebrates are sampled twice per year (summer and spring) at 51 sites on 22 rivers. These sites range from near pristine to those located in intensively farmed catchments, allowing numerous upstream-downstream comparisons; and impacted versus unimpacted comparisons and the major geological and ecoregions are included. Sites monitored for consents are included with SoE sites. The data provide a good picture of the ecological status of Taranaki rivers and is analysed for long term trends in river health based on the MCI score. Trends in MCI scores have been used by TRC to indicate where riparian management has been effective in improving river habitat condition (TRC 2009a).

Periphyton is sampled year per year (summer and spring) at 21 sites on 10 rivers. This regional coverage includes upper, middle and lower catchment sites and enables analysis of river slimes and trend analysis. Benthic cyanobacteria are sampled at 11 sites fortnightly during the period November-March.

3.2.3: Contact recreation and drinking water

Water samples are analysed for: faecal coliforms, E. coli and enterococci, 13 times a year during the period November to March at 14 popular freshwater bathing sites over summer. The annual medians of these data are used for trend analysis. A freshwater cyanobacteria monitoring programme sample five sites at popular freshwater bathing spots includes four lake beaches and one river site.

3.2.4: Groundwater

TRC participate in two national monitoring programmes that give resource information about their region. Five wells in the Taranki region are included in the IGNS national groundwater quality monitoring programme (NGMP). The NGMP analyses groundwater to 15 chemical constituents on a quarterly basis. The National Survey of Pesticides in Groundwater (NSPG) is conducted by ESR on a four yearly basis. The NSPG analyses for major pesticide groups include organonitrogen herbicide, acid herbicide and organochlorine pesticides. In addition, TRC monitor nitrogen in shallow groundwater at between 65 and 80 wells on a five-yearly basis. The groundwater programmes (Appendix 1) cover the major geological formations where aquifers occur and address the key points of resource use and availability, and quality in relation to land uses.

3.2.5: Lakes

Lake Rotorangi is only lake of significance in the Taranaki region. In our opinion the TRC monitoring programme adequately describes the chemical and nuisance plant status of the lake. Two sites on the lake are sampled four times annually for DRP, TN, conductivity, NOx-N, NH4-N, BOD, TP, clarity, temperature, plankton, benthic invertebrates. TRC carries out depth profiles of temperature and dissolved oxygen (DO) at three sites (the additional site is at the head of the lake) four times (seasonally) each year. Turbidity, black disc water clarity, suspended solids, conductivity, chlorophyll-a and pH are measured at all sites at the surface, and in the epilimnetic and hypolimnetic layers. Whole lake surveys of macrophytes are conducted every three years, and additional measurements are carried out to check for benthic nutrient release (occurring under temporarily deoxygenated conditions) during surveys in mid- to late summer.

3.2.6: QA/QC and data storage

All analyses are performed by the Taranki Regional Council IANZ registered chemistry laboratory using Standard Methods. The lab is not externally accredited for bacterial analysis, but performs QC/QA checks. Internal and external QC exercises for the SoE monitoring programme include an annual interlaboratory comparison where split samples are analysed by NIWA (Hamilton) and the results compared with the TRC laboratory. Recent results show good agreement (mostly +/-10%) between the two laboratories for a wide range of variables (TRC 2009b). In general, the TRC analytical laboratory performed well in interlaboratory QC tests.

All data from Taranaki Regional Council's field measurements and laboratory analyses are stored in a water quality archiving database which ensures the data can be efficiently extracted for analysis.

3.3: Horizons Regional Council

3.3.1: Rivers physical-chemical and microbiological

In common with some other regional councils, Horizons do not distinguish between SoE monitoring sites and other sites in their long-term monitoring programmes. For example, the current rivers SoE programme comprises 63 sites that are visited and sampled at monthly intervals (Table 4). Another 93 'reference' sites upstream of consented discharges and 13 contact recreation sites are monitored at regular intervals and are included in the overall monitoring programme to give a total of 169 sites that are used to describe the state of water quality in the region. Water quality data from monthly monitoring upstream of discharge sites is often amalgamated with SoE data because of the added information that is gained about water resource condition, although such sites are not called "SoE" sites. This has in the past led to some confusion. For example, a report by the Parliamentary Commissioner for the Environment (PCE, 2004) identified that many regional councils did not monitor BOD, although Horizons do monitor BOD at over 50 sites in the Manawatu catchment alone.

In the past, Horizons have employed a system of "rolling SoE sites" whereby some sites have been monitored on a rolling basis, i.e. once every three years 12 months of monthly sampling has been undertaken. This practice is no longer carried out by Horizons regional Council because of the need to have at least five years of continuous (monthly) monitoring data for trend analysis.

The 63 Horizons Regional Council SoE river sampling sites are located on 35 different rivers. The region has four major river systems: Whanganui, Rangatikei, Manawatu and Whangaehu that collectively have 17 sites as well as several tributaries with sampling sites. The current monitoring programme, therefore, covers a broad range of environmental conditions in the region. However, there were only 17 SoE water quality monitoring sites in the Horizons region that met our criteria for trend analysis. All but one of these sites was located in Low-elevation and Hill Topography and Pastoral Land-cover categories. These combinations of categories were also the most commonly occurring by river length (Table 7).

Table 7:

Numbers of river SoE water quality monitoring sites in the Horizons region that met our criteria for trend analysis and (in brackets) the number of apparent SoE sites in 2009 (i.e. having at least one sample in each season during 2009). The sites have been classified by REC Topography and Land-cover categories (see Table 3) and the colour scale indicates the total proportion of river length in the Region that is classified by each of the categories.

Horizons SoE monitoring samples are analysed for a suite of water quality variables but some detection levels appear high for the analysis of natural freshwaters. For example, the DRP detection limit was identified as too high for trend analysis by Ballantine & Davies-Colley (2009). Ammonium N is cited to 0.01 g m^-3, whereas the ANZECC (2000) default trigger value for ammonium N in moderately disturbed lowland rivers is 0.021 g m^-3 and some waters may have very low concentrations of c. 0.005 g m^-3. This will reduce the future ability of the program to detect trends in water bodies that are currently in good condition. Field measurements are made using established methods (e.g. calibrated field meters for dissolved oxygen (DO) and temperature). Some SoE sites have continuous monitoring of temperature (32), conductivity (3) and dissolved oxygen (5). Flow data (either measured or derived) are available for 54 sites but only 41 of these are SoE sites.

The small number of SoE sites in the Horizons region resulted in few statistically significant differences in water quality state for sites grouped by REC Topography and Land-cover categories (see Section 4.3.1). In addition, there were few detectable overall trends (in specific REC Topography and Land-cover categories) (see Section 4.3.2). These analyses indicate that the river water physical-chemical and microbiological quality monitoring program had insufficient statistical power (numbers of sites relative to their variability) to detect large scale patterns (as defined by REC categories) in water quality state and trends in the Horizons region. We note, however, that the statistical power of Horizon's SoE network will substantially increase in future now that the system of "rolling SoE sites" has been abandoned and the number of SoE sites in the monitoring network has been increased.

3.3.2: Rivers biological

Approximately 42 of the SoE sites are also surveyed annually for benthic invertebrates and monthly for periphyton. Results are expressed as macroinvertebrate community index (MCI) scores for state and trend analysis. Periphyton cover is estimated visually, but samples are also taken for chlorophyll a assessment of biomass.

3.3.3: Lakes

State-of-the-Environment lake monitoring is only conducted on Lake Horowhenua, but 3-4 lakes are sampled at weekly intervals during the bathing season for contact recreation. These samples are analysed for E. coli and cyanobacteria (which can sometimes be neurotoxic to humans and domestic animals on contact). Lake Horowhenua is sampled at monthly intervals at three sites along the axis of the lake. Samples are bulked and analysed as for river samples as wellas cyanobacteria during summer. We consider this monitoring programme is adequate for a shallow, dune lake such as Horowhenua.

3.3.4: Contact recreation and drinking water

Horizons Regional Council routinely tests water quality at fourteen popular swimming spots in the Region between 1 December and 30 April each summer season. Sites include coastal and river mouth beaches, rivers, Lake Wiritoa, Duddings Lake and Lake Horowhenua. Samples are analysed for cyanobacteria (blue-green algae) and faecal indicator organisms (Enterococci, faecal and total coliforms).

3.3.5: Groundwater

Seven groundwater management zones have been identified by HRC. Currently 32 groundwater sites are monitored of which 28 are SoE sites but since early 2008 this has been reduced to 25. All 25 sites are sampled once every seven months. This sampling frequency was considered the most cost-effective monitoring frequency in order to guarantee sampling of wells in all seasons in all months, with a return period of 12 years. On-site field measurements made at the time of sampling include temperature, electrical conductivity (EC), pH and redox potential. Laboratory analyses include major cations (Ca, Mg, Na, K) and anions (HCO₃, SO₄, Cl), important minor constituents (Fe, Mn), nutrients (NO_x-N, NH₄-NDRP), and other indicative parameters (SiO2, F, B, Br). Occasionally, samples are collected and analysed for microbial indicators, viz. E. Coli.

Horizons groundwater quality data is described in detail recent reviews by Daughney et al. (2009) and Zarour (2009). The main conclusions being that human activities have little detectable influence on groundwater quality and groundwater quality across the Region is similar to the average expectation for aquifers around the world. Trend tests for the sites that have sufficient data indicate that most parameters at most sites have remained constant with time. While it appears that groundwater quality in the Region has not been changing, there is a need to commit to long-term, regular groundwater quality monitoring at a core set of sites to confirm this assessment.

3.3.6: QA/QC and data storage

Samples were originally analysed by the Horizons Regional Council laboratory, which merged with the Palmerston City Council laboratory to form Central Environmental Laboratories (CEL) four years ago. Since 2009, samples have been analysed by Watercare Services Ltd., Auckland. Thus, the council has had analytical results from three different organisations with the prospect of incompatibility between the different analysts. The council have checked this by splitting samples and conducting inter-laboratory comparisons (the latest being a 3-laboratory comparison between CEL, Watercare and RJ Hills in 2010.

Benthic invertebrates are identified and enumerated by Stark Laboratories; with QA checks provided on 10% of samples by Biosorted Ltd. Five periphyton chlorophyll a samples a month being sent to NIWA, Christchurch, for QA comparison. All data from field measurements and laboratory analyses are stored in Horizons' water quality archiving database. This ensures the data can be efficiently extracted for analysis.

3.4: Environment Waikato

3.4.1: Rivers physical-chemical and microbiological

A total of 113 river sites in the Waikato region, including 10 Waikato River sites are currently being monitored and are reported on annually (Table 4). Data sets for many of the rivers began in 1990 although not all water quality variables have been monitored for the same length of time, e.g. records of visual clarity did not begin until 1995, while records of Escherichia coli did not begin until 1998.

Ten locations along the Waikato River are visited monthly (Taupo, Ohaaki, Ohakuri, Whakamaru, Waipapa, Hamilton-Narrows, Hamilton-Horotiu, Huntly, Mercer and Tuakau), and an additional four locations are included for the summer intensive microbiological survey. The major tributaries that enter the Waikato River are also monitored monthly as part of the Regional River Monitoring Programme (RERIMP) initiated in 1993. Three locations (Taupo at Reids Farm, Hamilton at Wellington Street, and Rangiriri) are sampled by NIWA as part of the NRWQN (EW 2008). The 13 Waikato River sites provide a comprehensive description of the Waikato River water quality along its length and, together with tributary river data and other monitoring information (e.g. from Consents to discharge wastewater), identify major inputs to the River.

The current level of monitoring and annual reporting of Waikato River water quality data seems comensurate with the importance of the Waikato River. However, the bottom waters of the Waikato River hydro-lakes are not monitored. The current EW main-stem sampling (10 sites) focuses only on the outflows from three of the lakes (i.e. the surface waters), effectively treating them as river sites. As a result little is known about any long-term changes in the bottom waters; particularly DO depletion and sediment release of As, N and P. We recognise though, that deepwater sampling would involve significant extra resources (boats, minimum of two staff per trip, specialised field instruments and samplers).

The monitoring of other Waikato rivers is based on the following major rivers and key tributaries: Upper Waikato River tributaries; Lower Waikato Tributaries; Waipa River; West Coast rivers, Coromandel Peninsula streams; Lake Taupo inflows; Hauraki Rivers (vis. Piako and Waihou River systems). These waters range in size and include approximately 20 sites where mean flows are < 1 m³/s and catchment areas < 30 km². The monitoring programme covers all major water resources within the region, giving a good description of the state of the Waikato regional rivers and enabling trend analysis to be conducted (e.g., Vant 2008).

There were 115 SoE water quality monitoring sites in the Waikato region that met our criteria for trend analysis. The majority of these sites were located in Low-elevation and Hill Topography and Pastoral Land-cover categories, which were also the most commonly occurring by river length (Table 8). However, the Exotic Forest, Indigenous Forest, Urban and Scrub Land-cover categories were also represented with reasonably balanced numbers of sites.

Table 8:

Numbers of river SoE water quality monitoring sites in the Waikato region that met our criteria for trend analysis and (in brackets) the number of apparent SoE sites in 2009 (i.e. having at least one sample in each season during 2009). The sites have been classified by REC Topography and Land-cover categories (see Table 3) and the colour scale indicates the total proportion of river length in the Region that is classified by each of the categories.

Water quality samples are analysed for up to 40 variables (27 routinely) for the purposes of ecological health, human uses including recreation, water supply and drinking water. The Waikato River Independent Scoping Study (WRISS) report has identified some issues relating to future iwi co-management of the Waikato River. WRISS identified the need for a cultural health index, to be developed by iwi that captures "health and wellbeing". The WRISS identified the need for fish and other kai abundance to be monitored. The WRISS report points out the difficulties in measuring the direct indicators (e.g. whitebait catch) and suggests some surrogates (e.g., habitat area). WRISS also identified the need for food safety assessment and possibly monitoring (e.g., mercury and arsenic in fish, faecal microbes on watercress, boron in drinking water, pesticides).

For 41 sites flow data is either available at or close to each site (primary) or is available from "secondary" sites (within about 20 km) enabling flow-adjustment of data for trend analysis (Vant 2008). For the remaining 72 sites a "flow index" is calculated, based on the flow at the time of sampling at a location elsewhere on the relevant river, or on a similar river nearby. This approach involves uncertainty in the flow estimates, with implications for trend analysis and for load calculations.

Statistically significant differences in water quality state for sites grouped by REC Topography and Land-cover categories were observed for most variables (see Section 4.4.1). In addition, overall trends (in specific REC Topography and Land-cover categories) were observed for several variables (see Section 4.4.2). These analyses indicate that the river water physical-chemical and microbiological quality monitoring program is adequate, at least from the perspective that it has sufficient statistical power (numbers of sites relative to their variability) to detect large scale patterns (as defined by REC categories) in water quality state and trends in the Waikato region.

3.4.2: Rivers biological

Environment Waikato has been carrying out annual assessments of invertebrate community composition in rivers since 1994 as part of the Regional Ecological Monitoring of Streams (REMS) programme. These sites include wadeable high-gradient rivers with stony beds, low-gradient wadeable streams dominated by soft sediments, and some larger non-wadeable streams with long term records that have been retained while sampling protocols are developed. From 2005, sampling has included a network of 23 wadeable 'reference sites' in undeveloped catchments to provide a baseline against which to measure change, and a range of sites representing low, moderate and high levels of pastoral land-cover ('land-cover sites'). Sampling at 46 'long-term sites', including three reference sites, has been conducted for more than 10 years using consistent protocols that have enabled assessment of temporal trends in ecological condition at these sites (Collier and Kelly 2005). Condition is assessed using four macroinvertebrate-based measures derived from 200+ counts of individuals: number of different types of mayflies, stoneflies and caddisflies (excluding algal-piercing Hydroptilidae)—EPT* richness; the percent abundance of these sensitive insects—%EPT; a measure of tolerance to organic pollution—the Macroinvertebrate Community Index or MCI; and an integrative score of all three metrics—Average Score Per Metric or ASPM. Metrics were also calculated reflecting (i) habitat quality based on qualitative assessments of nine riparian, bank and channel attributes, and (ii) instream plant cover and proliferation. Assessments of periphyton and macrophyte metrics were also made at most sites sampled since 2005, following the methods described in Collier et al. (2006).

The SoE monitoring of rivers by Environment Waikato is comprehensive and enables an overall understanding of the state of the region's rivers. The program has been in existence for a sufficiently long period for trends to be detected in relation to land-use. Some deficiencies in the program have been identified by EW staff, for example, ideally, the bottom-waters of the Waikato River hydro lakes would be routinely monitored and more sites would have well-rated flow gauges.

3.4.3: Rivers biological [sic]

3.4.4: Groundwater

A total of 11 groundwater level sites are monitored by Environment Waikato using permanent recorders. In 2008 a total of 550 samples were taken from 346 groundwater water quality sites as part of the following monitoring programmes: Regional Groundwater (110 wells annually), the National Groundwater Monitoring Programme (10 wells quarterly), Taupo (37 six-monthly) and Coromandel Groundwater Projects, Nitrates (30 quarterly), Pesticides (4 quarterly and 80 four-yearly) and Microbial (80 wells four-yearly). Community supplies are also monitored involving 88 schools biannually. The 110 regional wells are considered SoE sites and other nitrate, pesticide and microbial sites are subsets used for environmental indicators.

Regional (SoE) groundwater quality monitoring is undertaken annually with 23 variables determined for 110 sites. This network represents generally vulnerable aquifers with relatively young groundwater in aerobic conditions. Wells with significant iron concentrations are excluded from this network. Thirty sites are monitored quarterly and ten sites monitored quarterly as part of the National Groundwater Monitoring Programme. Routine water variables were determined on a quarterly basis at 37 sites as part of the Lake Taupo Project. In 2008, pesticides and microbial indicators were monitored at 80 sites across the region.

Comprehensive aquifer representation was a primary factor in site selection for the groundwater SoE programme. Aquifers likely to show temporal change were given preference over likely old waters and anaerobic conditions. Factors for individual well selection included good log information, ease of access and sampling.

Quarterly sampling is considered sufficiently frequent for changes in groundwater. Annual sampling reflects cost constraints. Lower frequency sampling e.g. four yearly for some indicators is used as a periodic check for any change rather than to detect linear trends.

Environment Waikato have an extensive groundwater monitoring SoE programme that represents all major aquifers in the region describes the groundwater resources, within the constraints of cost for this work. Monitoring is predominantly done annually and again reflects the cost limitations. Sampling protocols are in accordance with the National Protocol for SoE Groundwater Sampling (Daughney et al. 2006), and address issues of concern in the Waikato Region, notably nitrate.

3.4.5: Lakes

A long term programme monitoring the lake Taupo's water quality began in 1994. This program is conducted by NIWA with field assistance from the Department of Internal Affairs, Taupo Harbourmaster's Office. The programme initially focused on oxygen depletion rates, but now targets phytoplankton biomass, water clarity, and nutrient (particularly nitrate) accumulation in the lake. The long-term monitoring programme uses the historical mid-lake site, Site A. Monitoring of additional sites in the Kuratau Basin (Site B) and the Western Bays (Site C) between January 2002 and December 2004 suggested that spatial variability of water quality across Lake Taupo is minimal such that the mid-lake site may be regarded as representative of the open water quality of the lake. Further validation of the use of a single mid-lake monitoring site was obtained from a comparison of upper water column nutrient and chlorophyll a concentrations and algal enumeration between Site A and near-shore sites in Whangamata Bay (Kinloch) and Whakaipo Bay, over a two-year period from February 2007 up to June 2009. The study determined that "the near-shore water quality was very similar to the mid-lake water quality" and that "within this similarity in the measured data was much variability which may be due to short period time lags between the near-shore and mid-lake sites with respect to nutrient sources, and the zones of algal growth".

Water sampling is carried out at 2-4 week intervals and samples are analysed for Secchi disc, chlorophyll a, nitrate+nitrite-nitrogen, ammoniacal-N, dissolved organic N, particulate-N, total nitrogen, dissolved reactive phosphorus, dissolved organic phosphorus, particulate phosphorus, total phosphorus, and algal species dominance in integrated-tube water samples from the top 10 m. Zooplankton net hauls from 100 m (63 μm mesh) are preserved in 4% formalin and stored pending analysis. In addition, 12 bathing beaches around the lake are monitored weekly (up to 12 times) during summer (December-February), on alternate years, for water clarity and E. coli (a faecal indicator bacteria).

The other lakes in the Waikato region have been categorised by Environment Waikato into the following groups: Taupo volcanic zone lakes (viz. Lake Taupo); Waikato River hydro lakes (included in the Waikato River monitoring programme); peat lakes; riverine lakes; and west coast sand dune lakes. Environment Waikato currently monitors five shallow riverine lakes (Rotomanuka North, Rotoroa, Waahi, Waikare and Whangape). Monitoring of water quality follows the method established by the New Zealand Lakes Water Quality Monitoring Programme, which was subsequently adopted as a Ministry for the Environment protocol (Burns et al., 2000). Each site is monitored monthly, except for lakes Waahi, Waikare and Whangape, which are monitored every two months. Trophic level indicators are calculated for chlorophyll a, Secchi depth, total nitrogen and total phosphorus and results displayed on the EW website. Another four lakes (Mangahia, Ngaroto, Rotokauri and Rotomanuka South) had monitoring programmes that have now ceased because available information is deemed sufficient (Barnes 2002).

The importance of Lake Taupo and the current level of interest in preserving its near pristine state have provided the impetus for an intensive monitoring programme that meets both SoE needs and research on abating or mitigating adverse effects from land use activities. Most of the 100 or so lakes in the region are small (less than 10 ha) and monitoring is focused on those with ecological values or under land use pressures (e.g., eutrophication from pastoral runoff). West coast dune lakes are not included in the regular lake monitoring programme, but three of them (Harihari, Otamatearoa and Taharoa) are part of a group of lakes that have been studied as part of a "special investigation" to assess lake ecological health, including water quality, involving visits to each of lakes three times each year for the past four years (W. Vant, Environment Waikato. pers. comm.).

3.4.6: QA/QC and data storage

Chemical and microbial analyses of water samples are carried out with comprehensive QA/QC procedures. Some variables are measured in the field, but the majority are analysed by (IANZ-registered) Hill Laboratories using standard analytical methods with detection levels that are sufficiently low for environmental monitoring (EW 2008). Sample collection, stabilisation and transport are in accordance with ISO9001:2000 protocols for quality management. Back-up samples are held for two months until results have been verified by routine quality assurance procedures.

All data from field measurements and laboratory analyses are stored in Environment Waikato's water quality archiving database (TimeStudio). This ensures that data can be efficiently extracted for analysis.